90prime Dark Images in May/June 2004

C. W. Engelbracht; last updated 2004-7-8

Summary

Chip to Chip Variations

As shown in Table 1, the 90prime dark current is lowest on chips 1 and 2 and highest on chip 3. Chip 3 displays periodic structure in the column direction, while chip 4 displays a large region of elevated dark current in the center of the image.

Table 1 - Dark Variation from Chip to Chip
May 9 Combined 600s dark

Chip 1: display range -10 to 50
median=2.7
Chip 2: display range -10 to 50
median=7.9
Chip 3: display range 0 to 2000
median=783
Chip 4: display range -100 to 300
median=10.3

Variation With Time

As shown in Table 2, the structure of the dark current is similar from day to day but the level varies a bit, especially on chips 3 and 4.

Table 2 - Dark Measurements from Day to Day
May 10 Combined 600s dark

Chip 1: display range -10 to 50
median=3.4
Chip 2: display range -10 to 50
median=10.3
Chip 3: display range 0 to 2000
median=715
Chip 4: display range -100 to 300
median=12.1
May 11 Combined 600s dark

Chip 1: display range -10 to 50
median=2.5
Chip 2: display range -10 to 50
median=7.9
Chip 3: display range 0 to 2000
median=679
Chip 4: display range -100 to 300
median=9.2
June 20 Combined 600s dark

Chip 1: display range -10 to 50
median=2.1
Chip 2: display range -10 to 50
median=3.4
Chip 3: display range 0 to 2000
median=875
Chip 4: display range -100 to 300
median=9.5
June 21 Combined 600s dark

Chip 1: display range -10 to 50
median=1.7
Chip 2: display range -10 to 50
median=3.1
Chip 3: display range 0 to 2000
median=852
Chip 4: display range -100 to 300
median=7.9

The dark current on chips 3 and 4 varies on timescales as short as minutes. Table 3 shows plots of average dark current as a function of time for each chip during a sequence of dark observations. The points are separated by the duration of each dark measurement, i.e., 600 seconds. The dark level is observed to vary in a significant way on chips 3 and 4. The level of variability appears higher during the May run than the June run.

Table 3 - Dark Variability on Short Timescale
May 9 Individual 600s dark measurements

Chip 1
Chip 2
Chip 3
Chip 4
June 21 Individual 600s dark measurements

Chip 1
Chip 2
Chip 3
Chip 4

The dark level was observed to vary by up to 30% during the nights of June 20 and 21, 2004, but only on chip 4. This effect was not observing during the May 2004 run. The effect was corrected by scaling the dark current before subtracting it. Table 4 shows some nighttime data in the U band (600s exposures) on chips 3 and 4, the only ones with appreciable dark current. The standard (i.e., unscaled) dark subtraction works well on chip 3, but leaves obvious residuals on chip 4. Subtracting a scaled dark (reduced by 5% to 30% in the data shown below) corrects the dark subtraction for chip 4, but leaves obvious residuals in chip 3. Thus, to make nighttime flats for the June run, the dark was scaled for chip 4 only.

Table 4 - Nighttime Dark Variability in June 2004
Chip Standard Dark Scaled Dark
3
4

Variation With Integration Time

The dark current appears to scale properly with time, when measurements are made on the same day. The images in Table 5 are the differences of a scaled 600-second dark and a 60-second dark. The median values are all within a few tenths of an ADU of zero and no structure is observed in the images, so the dark appears to scale appropriately with time.

Table 5 - Difference of 60s and Scaled 600s Darks
(May 9 600s dark) / 10 - (May 9 60s dark)
display range is -20 to 20

Chip 1
Chip 2
Chip 3
Chip 4